We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
Meta: Llama 3.2 3B Instruct
meta-llama/llama-3.2-3b-instructTry Meta: Llama 3.2 3B Instruct now
Runs on your balance at the prices above. New accounts start with free usage.
One API. Every model.
Session cost $0.00
Settings
Advanced parameters
Meta: Llama 3.2 3B Instruct at a glance
- Input price is 89% lower than the median of 434 Text models.
- Context window of 131,072 tokens, enough for long documents and multi-turn conversations.
- Supports structured JSON output with a schema.
- A typical request (1,000 input / 500 output tokens) costs about $0.000764.
About Meta: Llama 3.2 3B Instruct
Meta: Llama 3.2 3B Instruct is a text model from Meta, identified on api14 as meta-llama/llama-3.2-3b-instruct. It is a 3-billion-parameter multilingual large language model built on a transformer architecture and tuned for instruction following. Meta released it on 25 September 2024, and its knowledge cutoff is 31 December 2023. The model sits in the text category, so it is aimed at language work rather than image, audio or other modalities.
On the input side it accepts text and returns text. The context window is 131,072 tokens, which is large for a model of this size and lets you pass long documents, transcripts or multi-turn histories in a single request. The maximum output is 117,964 tokens, so responses can be long as well. Meta describes the model as optimized for natural language processing tasks such as dialogue generation, reasoning and summarization, and its multilingual training means it can handle prompts in more than one language.
Those capabilities point to a fairly clear set of uses. Summarization fits naturally: long reports, meeting notes or article drafts can go in and a shorter version comes out. Dialogue generation makes it suitable for chat assistants, support bots and conversational interfaces where the instruction tuning keeps replies on task. Reasoning tasks, such as working through a problem stated in the prompt or extracting structured answers from unstructured text, are also in scope. Because the model is small, it is a reasonable choice when you want low latency or want to run many requests, and the long context means you do not have to chunk documents aggressively before sending them.
You call it through api14's OpenAI-compatible API using the model id meta-llama/llama-3.2-3b-instruct. Because the interface follows the OpenAI format, existing clients and SDKs generally work with only the base URL and model id changed. The request parameters it supports are frequency_penalty, logit_bias, logprobs, max_tokens, min_p, presence_penalty, repetition_penalty, seed, stop, structured_outputs, temperature, top_k, top_logprobs and top_p. That covers the usual sampling controls (temperature, top_p, top_k, min_p), penalties for repetition, deterministic runs via seed, stop sequences, token limits, and log probability inspection through logprobs and top_logprobs. Structured outputs are supported as well, which helps when you need the reply to conform to a schema.
The main thing to weigh is model size. At 3 billion parameters it is a compact model, so for tasks that demand deep reasoning or broad world knowledge, a larger model may serve you better. It is also text only, so anything involving other input types is out of scope here. If your workload is summarization, dialogue or moderate reasoning and you value a long context window, this model is a practical fit.
How to call the Meta: Llama 3.2 3B Instruct API
- Create an API key in your dashboard.
- Install the official openai SDK for your language.
- Set baseURL to https://api14.com/api/v1 and model to "meta-llama/llama-3.2-3b-instruct".
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "meta-llama/llama-3.2-3b-instruct",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
Supported parameters
frequency_penaltylogit_biaslogprobsmax_tokensmin_ppresence_penaltyrepetition_penaltyseedstopstructured_outputstemperaturetop_ktop_logprobstop_pSpecifications
- Context length
- 131,072
- Max output
- 72,000
- Input
- text
- Output
- text
- Knowledge cutoff
- 2023-12-31
Requests on api14 · last 7 days
Meta: Llama 3.2 3B Instruct: frequently asked questions
- How much does the Meta: Llama 3.2 3B Instruct API cost?
- Meta: Llama 3.2 3B Instruct costs Input $0.18 / 1M tokens. It is charged from your USD balance only for successful requests; failed requests are free.
- What is the context window of Meta: Llama 3.2 3B Instruct?
- Meta: Llama 3.2 3B Instruct reads up to 131,072 tokens per request: prompt, history and documents together. It writes up to 72,000 output tokens per response.
- What can Meta: Llama 3.2 3B Instruct do?
- Meta: Llama 3.2 3B Instruct accepts text and returns text. It supports structured JSON output.
- How do I call Meta: Llama 3.2 3B Instruct from my code?
- Point any OpenAI SDK at https://api14.com/api/v1 and set the model to "meta-llama/llama-3.2-3b-instruct". No other code changes. Copy a ready example from the code panel on this page.
- What are alternatives to Meta: Llama 3.2 3B Instruct?
- Comparable Text models at a similar price: Qwen: Qwen3 30B A3B Instruct 2507, Mistral: Mistral Small 4 (batch), ByteDance Seed: Seed 1.6 Flash. Each runs through the same API key, so switching is one string.
Meta: Llama 3.2 3B Instruct community
New topicThis community has no topics yet. Open the first one about Meta: Llama 3.2 3B Instruct.